Ikuya Yamada

Joint Learning of the Embedding of Words and Entities for Named Entity Disambiguation.

Named Entity Disambiguation (NED) refers to the task of resolving multiple named entity mentions in a document to their correct references in a knowledge base (KB) (e.g., Wikipedia). In this paper, we propose a novel embedding method specifically designed for NED. The proposed method jointly maps words and entities into the same continuous vector space. We extend the skip-gram model by using two models. The KB graph model learns the relatedness of entities using the link structure of the KB, whereas the anchor context model aims to align vectors such that similar words and entities occur close to one another in the vector space by leveraging KB anchors and their context words. By combining contexts based on the proposed embedding with standard NED features, we achieved state-of-the-art accuracy of 93.1% on the standard CoNLL dataset and 85.2% on the TAC 2010 dataset.

Enhancing Named Entity Recognition in Twitter Messages Using Entity Linking

In this paper, we describe our approach for Named Entity Recognition in Twitter, a shared task for ACL 2015 Workshop on Noisy User-generated Text (Baldwin et al., 2015). Because of the noisy, short, and colloquial nature of Twitter, the performance of Named Entity Recognition (NER) degrades significantly. To address this problem, we propose a novel method to enhance the performance of the Twitter NER task by using Entity Linking which is a method for detecting entity mentions in text and resolving them to corresponding entries in knowledge bases such as Wikipedia. Our method is based on supervised machine-learning and uses the highquality knowledge obtained from several open knowledge bases. In comparison with the other systems proposed for this shared task, our method achieved the best performance.

Named Entity Disambiguation for Noisy Text

We address the task of Named Entity Disambiguation (NED) for noisy text. We present WikilinksNED, a large-scale NED dataset of text fragments from the web, which is significantly noisier and more challenging than existing news-based datasets. To capture the limited and noisy local context surrounding each mention, we design a neural model and train it with a novel method for sampling informative negative examples. We also describe a new way of initializing word and entity embeddings that significantly improves performance. Our model significantly outperforms existing state-of-the-art methods on WikilinksNED while achieving comparable performance on a smaller newswire dataset.

Linkify: Enhancing Text Reading Experience by Detecting and Linking Helpful Entities to Users

We frequently encounter unfamiliar entity names (e.g., a persons name or a geographic location) while reading texts such as newspapers, magazines, and web pages. When this occurs, we typically perform a sequence of tedious actions: select the entity name, submit it to a search engine, and obtain detailed information from websites. In this paper, we present Linkify, a tool that enhances text reading by automatically converting entity names into links and displaying a widget that contains links to several relevant websites. We also propose a novel method for evaluating the helpfulness of entities to users using supervised machine learning with a set of carefully designed features. Experimental results show that our method significantly outperforms existing state-of-the-art methods.